# Carga de datos y pacqueterias necesarias.
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

# Librerías de Scikit-learn para el modelado
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder, StandardScaler
from sklearn.impute import SimpleImputer
from sklearn.compose import ColumnTransformer
from sklearn.pipeline import Pipeline
from sklearn.linear_model import LinearRegression
from sklearn.tree import DecisionTreeRegressor
from sklearn.ensemble import RandomForestRegressor
from sklearn.ensemble import GradientBoostingRegressor
from xgboost import XGBRegressor
from lightgbm import LGBMRegressor
import time
from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score

import warnings
warnings.filterwarnings('ignore')

# Carga del dataset
df_airbnb = pd.read_csv(r'C:\Users\Rogelio\Downloads\Ejercicio_Practico\Airbnb_Open_Data.csv', index_col='id')

#Cree una copia para las explicaciones
df_airbnb_copy = pd.read_csv(r'C:\Users\Rogelio\Downloads\Ejercicio_Practico\Airbnb_Open_Data.csv', index_col='id')

# Información general del DataFrame
df_airbnb_copy.info()
print(df_airbnb_copy.describe(include= 'all').round(2))

#Checar valores faltantes
print('Missing Values:')
print(df_airbnb_copy.isnull().mean().round(4) * 100)

#Checar valores duplicados
print('Duplicated Values:')
print(df_airbnb_copy.duplicated().sum())


df_airbnb_copy.head(5)

<class 'pandas.core.frame.DataFrame'>
Index: 102599 entries, 1001254 to 6094647
Data columns (total 25 columns):
 #   Column                          Non-Null Count   Dtype  
---  ------                          --------------   -----  
 0   NAME                            102349 non-null  object 
 1   host id                         102599 non-null  int64  
 2   host_identity_verified          102310 non-null  object 
 3   host name                       102193 non-null  object 
 4   neighbourhood group             102570 non-null  object 
 5   neighbourhood                   102583 non-null  object 
 6   lat                             102591 non-null  float64
 7   long                            102591 non-null  float64
 8   country                         102067 non-null  object 
 9   country code                    102468 non-null  object 
 10  instant_bookable                102494 non-null  object 
 11  cancellation_policy             102523 non-null  object 
 12  room type                       102599 non-null  object 
 13  Construction year               102385 non-null  float64
 14  price                           102352 non-null  object 
 15  service fee                     102326 non-null  object 
 16  minimum nights                  102190 non-null  float64
 17  number of reviews               102416 non-null  float64
 18  last review                     86706 non-null   object 
 19  reviews per month               86720 non-null   float64
 20  review rate number              102273 non-null  float64
 21  calculated host listings count  102280 non-null  float64
 22  availability 365                102151 non-null  float64
 23  house_rules                     50468 non-null   object 
 24  license                         2 non-null       object 
dtypes: float64(9), int64(1), object(15)
memory usage: 20.4+ MB
                       NAME       host id host_identity_verified host name  \
count                102349  1.025990e+05                 102310    102193   
unique                61281           NaN                      2     13190   
top     Home away from home           NaN            unconfirmed   Michael   
freq                     33           NaN                  51200       881   
mean                    NaN  4.925411e+10                    NaN       NaN   
std                     NaN  2.853900e+10                    NaN       NaN   
min                     NaN  1.236005e+08                    NaN       NaN   
25%                     NaN  2.458333e+10                    NaN       NaN   
50%                     NaN  4.911774e+10                    NaN       NaN   
75%                     NaN  7.399650e+10                    NaN       NaN   
max                     NaN  9.876313e+10                    NaN       NaN   

       neighbourhood group       neighbourhood        lat       long  \
count               102570              102583  102591.00  102591.00   
unique                   7                 224        NaN        NaN   
top              Manhattan  Bedford-Stuyvesant        NaN        NaN   
freq                 43792                7937        NaN        NaN   
mean                   NaN                 NaN      40.73     -73.95   
std                    NaN                 NaN       0.06       0.05   
min                    NaN                 NaN      40.50     -74.25   
25%                    NaN                 NaN      40.69     -73.98   
50%                    NaN                 NaN      40.72     -73.95   
75%                    NaN                 NaN      40.76     -73.93   
max                    NaN                 NaN      40.92     -73.71   

              country country code  ... service fee minimum nights  \
count          102067       102468  ...      102326      102190.00   
unique              1            1  ...         231            NaN   
top     United States           US  ...        $41             NaN   
freq           102067       102468  ...         526            NaN   
mean              NaN          NaN  ...         NaN           8.14   
std               NaN          NaN  ...         NaN          30.55   
min               NaN          NaN  ...         NaN       -1223.00   
25%               NaN          NaN  ...         NaN           2.00   
50%               NaN          NaN  ...         NaN           3.00   
75%               NaN          NaN  ...         NaN           5.00   
max               NaN          NaN  ...         NaN        5645.00   

       number of reviews  last review reviews per month review rate number  \
count          102416.00        86706          86720.00          102273.00   
unique               NaN         2477               NaN                NaN   
top                  NaN    6/23/2019               NaN                NaN   
freq                 NaN         2443               NaN                NaN   
mean               27.48          NaN              1.37               3.28   
std                49.51          NaN              1.75               1.28   
min                 0.00          NaN              0.01               1.00   
25%                 1.00          NaN              0.22               2.00   
50%                 7.00          NaN              0.74               3.00   
75%                30.00          NaN              2.00               4.00   
max              1024.00          NaN             90.00               5.00   

        calculated host listings count  availability 365 house_rules   license  
count                        102280.00         102151.00       50468         2  
unique                             NaN               NaN        1976         1  
top                                NaN               NaN      #NAME?  41662/AL  
freq                               NaN               NaN        2712         2  
mean                              7.94            141.13         NaN       NaN  
std                              32.22            135.44         NaN       NaN  
min                               1.00            -10.00         NaN       NaN  
25%                               1.00              3.00         NaN       NaN  
50%                               1.00             96.00         NaN       NaN  
75%                               2.00            269.00         NaN       NaN  
max                             332.00           3677.00         NaN       NaN  

[11 rows x 25 columns]
Missing Values:
NAME                                0.24
host id                             0.00
host_identity_verified              0.28
host name                           0.40
neighbourhood group                 0.03
neighbourhood                       0.02
lat                                 0.01
long                                0.01
country                             0.52
country code                        0.13
instant_bookable                    0.10
cancellation_policy                 0.07
room type                           0.00
Construction year                   0.21
price                               0.24
service fee                         0.27
minimum nights                      0.40
number of reviews                   0.18
last review                        15.49
reviews per month                  15.48
review rate number                  0.32
calculated host listings count      0.31
availability 365                    0.44
house_rules                        50.81
license                           100.00
dtype: float64
Duplicated Values:
541

#Encontrar Duplicados
id_duplicated_rows = df_airbnb_copy[df_airbnb_copy.duplicated(keep=False)]
id_duplicated_rows_sorted = id_duplicated_rows.sort_values(by='id')
id_duplicated_rows_sorted

# Eliminacion de columnas irrelevantes
cols_to_drop_if_exist = ['host name', 'country', 'country code', 'house_rules', 'license']
cols_to_drop = [col for col in cols_to_drop_if_exist if col in df_airbnb.columns]
if cols_to_drop:
    df_airbnb_cleaned = df_airbnb.drop(cols_to_drop, axis=1)
else:
    df_airbnb_cleaned = df_airbnb

#Cambiando columnas price y service fee a numericas
if 'price' in df_airbnb_cleaned.columns:
    df_airbnb_cleaned['price'] = df_airbnb_cleaned['price'].replace({r'\$': '', ',': ''}, regex=True).astype(float)
if 'service fee' in df_airbnb_cleaned.columns:
    df_airbnb_cleaned['service fee'] = df_airbnb_cleaned['service fee'].str.replace(r'\$', '', regex=True).str.replace(',', '', regex=True).str.strip().astype(float)

#Estandarizando nombres de columnas
df_airbnb_cleaned.columns = df_airbnb_cleaned.columns.str.strip().str.lower().str.replace(' ', '_').str.replace('(', '').str.replace(')', '')

print(df_airbnb_cleaned['host_identity_verified'].unique())
print(df_airbnb_cleaned['neighbourhood_group'].unique()) 
print(df_airbnb_cleaned['cancellation_policy'].unique()) 
print(df_airbnb_cleaned['room_type'].unique())

['unconfirmed' 'verified' nan]
['Brooklyn' 'Manhattan' 'brookln' 'manhatan' 'Queens' nan 'Staten Island'
 'Bronx']
['strict' 'moderate' 'flexible' nan]
['Private room' 'Entire home/apt' 'Shared room' 'Hotel room']

# Mapeo correcto de valores categoricos
correccion_neighbourhood_group = {
    'brookln': 'Brooklyn',
    'Brookln' : 'Brooklyn',
    'Brooklyn': 'Brooklyn',
    'manhatan': 'Manhattan',
    'Manhatan': 'Manhattan',
    'Manhattan': 'Manhattan'}

# Limpieza
df_airbnb_cleaned['neighbourhood_group'] = (
    df_airbnb_cleaned['neighbourhood_group']
    .str.strip()
    .str.title()
    .replace(correccion_neighbourhood_group)           # Aplica correcciones
)

# Normalizando todos los nombres de barrios
df_airbnb_cleaned['neighbourhood'] = (
    df_airbnb_cleaned['neighbourhood']
    .astype(str)
    .str.strip()
    .str.lower()
    .str.replace("’", "'")
    .str.replace(r"[^\w\s']", '', regex=True)
    .str.title()
)

#Eliminando duplicados
df_airbnb_cleaned = df_airbnb_cleaned.drop_duplicates()

#Missing values
df_airbnb_cleaned['name'] = df_airbnb_cleaned['name'].fillna('No Name')

df_airbnb_cleaned['host_identity_verified'] = df_airbnb_cleaned['host_identity_verified'].fillna('unconfirmed')

#Ajustando valores faltantes
df_airbnb_cleaned['neighbourhood_group'] = df_airbnb_cleaned['neighbourhood_group'].fillna('Other')

df_airbnb_cleaned['neighbourhood'] = df_airbnb_cleaned['neighbourhood'].fillna('Unknown')

df_airbnb_cleaned['instant_bookable'] = df_airbnb_cleaned['instant_bookable'].fillna(df_airbnb_cleaned['instant_bookable'].mode()[0])

df_airbnb_cleaned['cancellation_policy'] = df_airbnb_cleaned['cancellation_policy'].fillna(df_airbnb_cleaned['cancellation_policy'].mode()[0])

df_airbnb_cleaned['availability_365'] = df_airbnb_cleaned['availability_365'].fillna(0)

df_airbnb_cleaned['service_fee'] = df_airbnb_cleaned['service_fee'].fillna(0)

df_airbnb_cleaned['number_of_reviews'] = df_airbnb_cleaned['number_of_reviews'].fillna(0)


df_airbnb_cleaned.dropna(subset=['lat', 'long'], inplace=True)

median_reviews = df_airbnb_cleaned['reviews_per_month'].median()
df_airbnb_cleaned['reviews_per_month'].fillna(median_reviews, inplace=True)

median_host_count = df_airbnb_cleaned['calculated_host_listings_count'].median()
df_airbnb_cleaned['calculated_host_listings_count'].fillna(median_host_count, inplace=True)

median_rate = df_airbnb_cleaned['review_rate_number'].median()
df_airbnb_cleaned['review_rate_number'].fillna(median_rate, inplace=True)

df_airbnb_cleaned['construction_year'] = df_airbnb_cleaned['construction_year'].fillna(df_airbnb_cleaned['construction_year'].median())

df_airbnb_cleaned.dropna(subset=['price','minimum_nights'], inplace=True)

# Reemplazar valores negativos con abs
df_airbnb_cleaned['minimum_nights'] = df_airbnb_cleaned['minimum_nights'].apply(lambda x: x if x > 1 else abs(x))

#Rememplazar valores negativos con abs
df_airbnb_cleaned['availability_365'] = df_airbnb_cleaned['availability_365'].apply(lambda x: x if x > 1 else abs(x))
# Asegurar que los valores estén en el rango de 0 a 365
df_airbnb_cleaned['availability_365'] = df_airbnb_cleaned['availability_365'].clip(lower=1, upper=365)
df_airbnb_cleaned['availability_365'] = df_airbnb_cleaned['availability_365'].astype(int)


# Conviertiendo a fechas
if 'last_review' in df_airbnb_cleaned.columns:
    df_airbnb_cleaned['last_review'] = pd.to_datetime(df_airbnb_cleaned['last_review'], errors='coerce')

# Convertir a categorias
for col in ['host_identity_verified', 'neighbourhood_group', 'neighbourhood', 'instant_bookable', 'cancellation_policy', 'room_type']:
     if col in df_airbnb_cleaned.columns:
            df_airbnb_cleaned[col] = df_airbnb_cleaned[col].astype('category')

#convertir a int
df_airbnb_cleaned['construction_year'] = df_airbnb_cleaned['construction_year'].astype('Int64')
df_airbnb_cleaned['number_of_reviews'] = df_airbnb_cleaned['number_of_reviews'].astype(int)
df_airbnb_cleaned['minimum_nights'] = df_airbnb_cleaned['minimum_nights'].astype(int)

# Análisis final del DataFrame limpio

df_airbnb_cleaned = df_airbnb_cleaned.sort_index()
df_airbnb_cleaned.info()
print(df_airbnb_cleaned.describe(include= 'all').round(2))

print('duplicated values')
print(df_airbnb_cleaned.duplicated().sum())

print('missing values')
print(df_airbnb_cleaned.isnull().sum().sort_values(ascending=False))

display(df_airbnb_cleaned.head(5))

<class 'pandas.core.frame.DataFrame'>
Index: 101403 entries, 1001254 to 57367417
Data columns (total 20 columns):
 #   Column                          Non-Null Count   Dtype         
---  ------                          --------------   -----         
 0   name                            101403 non-null  object        
 1   host_id                         101403 non-null  int64         
 2   host_identity_verified          101403 non-null  category      
 3   neighbourhood_group             101403 non-null  category      
 4   neighbourhood                   101403 non-null  category      
 5   lat                             101403 non-null  float64       
 6   long                            101403 non-null  float64       
 7   instant_bookable                101403 non-null  category      
 8   cancellation_policy             101403 non-null  category      
 9   room_type                       101403 non-null  category      
 10  construction_year               101403 non-null  Int64         
 11  price                           101403 non-null  float64       
 12  service_fee                     101403 non-null  float64       
 13  minimum_nights                  101403 non-null  int64         
 14  number_of_reviews               101403 non-null  int64         
 15  last_review                     85659 non-null   datetime64[ns]
 16  reviews_per_month               101403 non-null  float64       
 17  review_rate_number              101403 non-null  float64       
 18  calculated_host_listings_count  101403 non-null  float64       
 19  availability_365                101403 non-null  int64         
dtypes: Int64(1), category(6), datetime64[ns](1), float64(7), int64(4), object(1)
memory usage: 12.4+ MB
           name       host_id host_identity_verified neighbourhood_group  \
count    101403  1.014030e+05                 101403              101403   
unique    60895           NaN                      2                   6   
top     No Name           NaN            unconfirmed           Manhattan   
freq        241           NaN                  50908               43257   
mean        NaN  4.926152e+10                    NaN                 NaN   
min         NaN  1.236005e+08                    NaN                 NaN   
25%         NaN  2.458389e+10                    NaN                 NaN   
50%         NaN  4.912666e+10                    NaN                 NaN   
75%         NaN  7.400226e+10                    NaN                 NaN   
max         NaN  9.876313e+10                    NaN                 NaN   
std         NaN  2.853789e+10                    NaN                 NaN   

            neighbourhood        lat       long instant_bookable  \
count              101403  101403.00  101403.00           101403   
unique                225        NaN        NaN                2   
top     Bedfordstuyvesant        NaN        NaN            False   
freq                 7852        NaN        NaN            50964   
mean                  NaN      40.73     -73.95              NaN   
min                   NaN      40.50     -74.25              NaN   
25%                   NaN      40.69     -73.98              NaN   
50%                   NaN      40.72     -73.95              NaN   
75%                   NaN      40.76     -73.93              NaN   
max                   NaN      40.92     -73.71              NaN   
std                   NaN       0.06       0.05              NaN   

       cancellation_policy        room_type  construction_year      price  \
count               101403           101403           101403.0  101403.00   
unique                   3                4               <NA>        NaN   
top               moderate  Entire home/apt               <NA>        NaN   
freq                 34027            53073               <NA>        NaN   
mean                   NaN              NaN            2012.49     625.40   
min                    NaN              NaN             2003.0      50.00   
25%                    NaN              NaN             2008.0     340.00   
50%                    NaN              NaN             2012.0     625.00   
75%                    NaN              NaN             2017.0     913.00   
max                    NaN              NaN             2022.0    1200.00   
std                    NaN              NaN               5.76     331.69   

        service_fee  minimum_nights  number_of_reviews  \
count     101403.00       101403.00          101403.00   
unique          NaN             NaN                NaN   
top             NaN             NaN                NaN   
freq            NaN             NaN                NaN   
mean         124.77            8.16              27.43   
min            0.00            1.00               0.00   
25%           68.00            2.00               1.00   
50%          125.00            3.00               7.00   
75%          182.00            5.00              30.00   
max          240.00         5645.00            1024.00   
std           66.54           30.61              49.44   

                          last_review  reviews_per_month  review_rate_number  \
count                           85659          101403.00           101403.00   
unique                            NaN                NaN                 NaN   
top                               NaN                NaN                 NaN   
freq                              NaN                NaN                 NaN   
mean    2019-06-10 18:24:59.611249408               1.28                3.28   
min               2012-07-11 00:00:00               0.01                1.00   
25%               2018-10-27 00:00:00               0.28                2.00   
50%               2019-06-13 00:00:00               0.74                3.00   
75%               2019-07-05 00:00:00               1.71                4.00   
max               2058-06-16 00:00:00              90.00                5.00   
std                               NaN               1.62                1.28   

        calculated_host_listings_count  availability_365  
count                        101403.00         101403.00  
unique                             NaN               NaN  
top                                NaN               NaN  
freq                               NaN               NaN  
mean                              7.90            139.74  
min                               1.00              1.00  
25%                               1.00              3.00  
50%                               1.00             95.00  
75%                               2.00            268.00  
max                             332.00            365.00  
std                              32.18            133.20  
duplicated values
0
missing values
last_review                       15744
name                                  0
host_identity_verified                0
host_id                               0
neighbourhood_group                   0
neighbourhood                         0
long                                  0
lat                                   0
cancellation_policy                   0
room_type                             0
construction_year                     0
instant_bookable                      0
price                                 0
service_fee                           0
minimum_nights                        0
number_of_reviews                     0
reviews_per_month                     0
review_rate_number                    0
calculated_host_listings_count        0
availability_365                      0
dtype: int64

# Lista de las columnas numéricas a visualizar

numeric_cols = [
    'price', 'service_fee', 'minimum_nights', 'number_of_reviews',
    'reviews_per_month', 'calculated_host_listings_count',
    'review_rate_number', 'availability_365'
]


fig, axes = plt.subplots(2, 4, figsize=(20, 10))
fig.suptitle('Análisis de Outliers en Variables Numéricas', fontsize=20, y=1.02)

axes = axes.flatten()

for i, col in enumerate(numeric_cols):
    sns.boxplot(x=df_airbnb_cleaned[col], ax=axes[i], color='skyblue')
    axes[i].set_title(col, fontsize=12)
    axes[i].set_xlabel('')

plt.tight_layout()
plt.show()

from scipy.stats import skew

numerical_cols = df_airbnb_cleaned.select_dtypes(include=np.number).columns

skewness = df_airbnb_cleaned[numerical_cols].skew().sort_values(ascending=False)

print("Coeficiente de Asimetría para todas las columnas numéricas:")
print("-" * 50)
print(skewness)

Coeficiente de Asimetría para todas las columnas numéricas:
--------------------------------------------------
minimum_nights                    88.167141
reviews_per_month                  7.616094
calculated_host_listings_count     7.239947
number_of_reviews                  3.838713
long                               1.241487
availability_365                    0.45869
lat                                0.230168
construction_year                   0.00637
host_id                            0.006234
price                              0.000822
service_fee                       -0.001106
review_rate_number                 -0.13769
dtype: Float64

# Estudio de las variables lat y long
fig, axes = plt.subplots(2, 2, figsize=(14, 8))

# latitud
sns.boxplot(data=df_airbnb_cleaned, x='lat', ax=axes[0,0])
axes[0,0].set_title('Boxplot de Latitud')

sns.histplot(df_airbnb_cleaned['lat'].dropna(), kde=True, ax=axes[0,1])
axes[0,1].set_title('Histograma de Latitud')

# longitud
sns.boxplot(data=df_airbnb_cleaned, x='long', ax=axes[1,0])
axes[1,0].set_title('Boxplot de Longitud')

sns.histplot(df_airbnb_cleaned['long'].dropna(), kde=True, ax=axes[1,1])
axes[1,1].set_title('Histograma de Longitud')

plt.tight_layout()
plt.show()

plt.figure(figsize=(8,4))
sns.boxplot(x=df_airbnb_cleaned['minimum_nights'])
plt.title('Boxplot de minimum_nights')
plt.show()

# limite suporior para outliers

Q1 = df_airbnb_cleaned['minimum_nights'].quantile(0.25)
Q3 = df_airbnb_cleaned['minimum_nights'].quantile(0.75)
IQR = Q3 - Q1
upper_bound = Q3 + 1.5 * IQR

print(f"Q3: {Q3}, IQR: {IQR}, upper_bound: {upper_bound}")

Q3: 5.0, IQR: 3.0, upper_bound: 9.5

# Filtrado del DataFrame para quedarme solo con las filas donde minimum_nights es menor o igual al límite superior calculado
df_airbnb_cleaned = df_airbnb_cleaned[df_airbnb_cleaned['minimum_nights'] <= upper_bound]

# Hacer una copia explícita del DataFrame luego de filtrar, elimina la ambigüedad
df_airbnb_cleaned = df_airbnb_cleaned[df_airbnb_cleaned['minimum_nights'] <= upper_bound].copy()

df_asimetrico = df_airbnb_cleaned.copy()   # copia para comparar resulatdos en una grafica posterior

# Definir los límites de capping basados en el contexto del negocio
cap_reviews_per_month = 20  # Más de 20 reseñas por mes es inusual
cap_calculated_host_listings_count = 100 # Más de 100 listados es inusual
cap_number_of_reviews = 500  # Más de 500 reseñas es un valor extremo

# capping
df_airbnb_cleaned['reviews_per_month'] = df_airbnb_cleaned['reviews_per_month'].clip(upper=cap_reviews_per_month)
df_airbnb_cleaned['calculated_host_listings_count'] = df_airbnb_cleaned['calculated_host_listings_count'].clip(upper=cap_calculated_host_listings_count)
df_airbnb_cleaned['number_of_reviews'] = df_airbnb_cleaned['number_of_reviews'].clip(upper=cap_number_of_reviews)

# recalculo de asimetria
numerical_cols = df_airbnb_cleaned.select_dtypes(include=np.number).columns
skewness_new = df_airbnb_cleaned[numerical_cols].skew().sort_values(ascending=False)

print("Coeficiente de Asimetría después del Capping:")
print("-" * 50)
print(skewness_new)

Coeficiente de Asimetría después del Capping:
--------------------------------------------------
calculated_host_listings_count    8.369815
number_of_reviews                 3.311504
reviews_per_month                 2.479984
minimum_nights                    1.209769
long                              1.194542
availability_365                   0.57565
lat                               0.279667
construction_year                 0.006147
host_id                           0.006012
price                             0.000647
service_fee                      -0.000997
review_rate_number               -0.136041
dtype: Float64

# Lista de columnas que aún necesitan una transformación
columns_to_transform = [
    'calculated_host_listings_count',
    'number_of_reviews',
    'reviews_per_month',
    'minimum_nights'
]

# transformación logarítmica 
for col in columns_to_transform:
    df_airbnb_cleaned[col] = np.log1p(df_airbnb_cleaned[col])

# Resultado final de asimetria
numerical_cols = df_airbnb_cleaned.select_dtypes(include=np.number).columns

skewness = df_airbnb_cleaned[numerical_cols].skew().sort_values(ascending=False)

print("Coeficiente de Asimetría para todas las columnas numéricas:")
print("-" * 50)
print(skewness)

Coeficiente de Asimetría para todas las columnas numéricas:
--------------------------------------------------
calculated_host_listings_count    3.259532
long                              1.194542
reviews_per_month                 0.726478
availability_365                   0.57565
minimum_nights                    0.439075
lat                               0.279667
number_of_reviews                 0.126472
construction_year                 0.006147
host_id                           0.006012
price                             0.000647
service_fee                      -0.000997
review_rate_number               -0.136041
dtype: Float64

fig, axes = plt.subplots(nrows=len(columns_to_transform), ncols=2, figsize=(15, 12))
fig.suptitle('Distribución de Columnas Clave: Antes y Después de la Transformación Logarítmica', fontsize=18, y=1.02)

for i, col in enumerate(columns_to_transform):
    # Gráfico 'Antes'
    sns.histplot(data=df_asimetrico, x=col, bins=50, kde=True, ax=axes[i, 0], color='salmon')
    axes[i, 0].set_title(f'Antes: {col}\nAsimetría: {df_asimetrico[col].skew():.2f}', fontsize=12)
    axes[i, 0].set_xlabel('')
    axes[i, 0].set_ylabel('Frecuencia')

    # Gráfico 'Después'
    sns.histplot(data=df_airbnb_cleaned, x=col, bins=50, kde=True, ax=axes[i, 1], color='royalblue')
    axes[i, 1].set_title(f'Después: {col}\nAsimetría: {df_airbnb_cleaned[col].skew():.2f}', fontsize=12)
    axes[i, 1].set_xlabel('')
    axes[i, 1].set_ylabel('')

plt.tight_layout()
plt.show()

# Mapa de Calor

numerical_cols = ['price', 'service_fee', 'minimum_nights', 'number_of_reviews',
                  'reviews_per_month', 'review_rate_number',
                  'calculated_host_listings_count', 'availability_365',
                  'lat', 'long']

correlation_matrix = df_airbnb_cleaned[numerical_cols].corr()

plt.figure(figsize=(12, 10))
sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm', fmt=".2f", linewidths=.5)
plt.title('Mapa de Calor de la Matriz de Correlación de Variables Numéricas', fontsize=16)
plt.show()

# Precio por tipo de habitación

plt.figure(figsize=(10, 6))
sns.boxplot(x='room_type', y='price', data=df_airbnb_cleaned)

plt.title('Distribución de Precios por Tipo de Habitación')
plt.xlabel('Tipo de Habitación')
plt.ylabel('Precio por Noche ($)')

plt.show()

# barrios por precio promedio
neighborhood_order = df_airbnb_cleaned.groupby('neighbourhood_group')['price'].median().sort_values(ascending=False).index

plt.figure(figsize=(12, 7))
sns.boxplot(x='neighbourhood_group', y='price', data=df_airbnb_cleaned, order=neighborhood_order)

plt.title('Distribución de Precios por Grupo de Barrio')
plt.xlabel('Grupo de Barrio')
plt.ylabel('Precio por Noche ($)')
plt.xticks(rotation=45) 

plt.show()

# Distribución de Precios en Bins de $100
frecuencia_por_rango = pd.Series([4350, 8701, 8820, 8622, 9050, 8923, 8906, 8786, 8647, 8854, 8853, 8823],
                                 index=['0-100', '100-200', '200-300', '300-400', '400-500', '500-600', '600-700', '700-800', '800-900', '900-1000', '1000-1100', '1100-1200'])

frecuencia_por_rango.plot(kind='bar', figsize=(15, 8), color='darkred')

plt.title('Distribución de Precios de Alojamientos (en Bins de $100)', fontsize=16)
plt.xlabel('Rango de Precio ($)', fontsize=12)
plt.ylabel('Cantidad de Alojamientos', fontsize=12)
plt.xticks(rotation=45)
plt.show()

# Frecuencia y Distribución por Tipo de Habitación

values = df_airbnb_cleaned['room_type'].value_counts()
keys = values.index
total_records = values.sum()
colors = sns.color_palette("plasma", len(keys))

fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(16, 8))
fig.suptitle('Frecuencia y Distribución por Tipo de Habitación', fontsize=20, fontweight='bold', color='darkblue')

# --- Gráfico de Barras ---
sns.countplot(data=df_airbnb_cleaned, x='room_type', palette=colors, ax=ax1, order=keys)
ax1.set_title('Conteo de Propiedades por Tipo', fontsize=16)
ax1.set_xlabel('Tipo de Habitación', fontsize=12)
ax1.set_ylabel('Cantidad de Propiedades', fontsize=12)
ax1.grid(axis='y', linestyle='', alpha=0.7)

for p in ax1.patches:
    height = p.get_height()
    ax1.annotate(f'{height:,}',
                  (p.get_x() + p.get_width() / 2., height),
                  ha='center', va='bottom', fontsize=12, fontweight='bold')

# --- Gráfico Circular tipo Donut ---
labels = [f"{k} - {v/total_records*100:.2f}%" for k, v in zip(keys, values)]
explode = [0.05] * len(keys)

patches, texts = ax2.pie(values, startangle=90, explode=explode, colors=colors)
centre_circle = plt.Circle((0, 0), 0.70, fc='white')
ax2.add_artist(centre_circle)
ax2.set_title('Proporción por Tipo de Habitación', fontsize=16)
ax2.axis('equal')
ax2.legend(patches, labels, loc='center left', bbox_to_anchor=(1.1, 0.5), fontsize=10)

plt.tight_layout(rect=[0, 0.03, 1, 0.95])
plt.show()

# Alojamientos por Rango de Precio y Ubicación

df_airbnb_cleaned['price_range'] = pd.cut(df_airbnb_cleaned['price'], bins=[0, 100, 200, 300, 500, np.inf],
                           labels=['<100', '100-200', '200-300', '300-500', '>500'])

plt.figure(figsize=(15, 8))
sns.countplot(x='neighbourhood_group', hue='price_range', data=df_airbnb_cleaned)
plt.title('Segmentación de Alojamientos por Rango de Precio y Ubicación')
plt.xlabel('Grupo de Vecindario')
plt.ylabel('Número de Alojamientos')
plt.xticks(rotation=45)
plt.legend(title='Rango de Precio')
plt.show()

# Alojamientos por Tipo de Habitación y Ubicación

plt.figure(figsize=(12, 7))
sns.countplot(data=df_airbnb_cleaned, x='neighbourhood_group', hue='room_type', palette='viridis')
plt.title('Distribución de Tipos de Habitación por Grupo de Barrio', fontsize=16)
plt.xlabel('Grupo de Barrio')
plt.ylabel('Cantidad de Alojamientos')
plt.xticks(rotation=45)
plt.show()

import geopandas as gpd
import contextily as cx

gdf = gpd.GeoDataFrame(
    df_airbnb_cleaned,
    geometry=gpd.points_from_xy(df_airbnb_cleaned['long'], df_airbnb_cleaned['lat']),
    crs="EPSG:4326"
)

fig, ax = plt.subplots(figsize=(14, 12))

gdf.plot(
    column='price',
    ax=ax,
    legend=True,
    legend_kwds={'label': "Precio por Noche ($)"},
    cmap='viridis',
    markersize=10,
    alpha=0.6
)

cx.add_basemap(ax, crs=gdf.crs.to_string(), source=cx.providers.CartoDB.Positron)

ax.set_title('Distribución Geográfica de Precios', fontsize=18)
ax.set_xlabel('Longitud')
ax.set_ylabel('Latitud')

plt.show()

# Alojamientos por Grupo de Barrio y Ubicación

gdf = gpd.GeoDataFrame(
    df_airbnb_cleaned,
    geometry=gpd.points_from_xy(df_airbnb_cleaned['long'], df_airbnb_cleaned['lat']),
    crs="EPSG:4326"
)

fig, ax = plt.subplots(figsize=(14, 12))

gdf.plot(
    column='neighbourhood_group',
    ax=ax,
    legend=True,
    legend_kwds={'title': "Grupo de Barrio"},
    cmap='viridis',
    markersize=10,
    alpha=0.6
)

cx.add_basemap(
    ax,
    crs=gdf.crs.to_string(),
    source=cx.providers.CartoDB.Positron
)

ax.set_title('Distribución Geográfica de Alojamientos por Grupo de Barrio', fontsize=18)
ax.set_xlabel('Longitud')
ax.set_ylabel('Latitud')

plt.show()

# Reseñas por Ubicación
gdf = gpd.GeoDataFrame(
    df_airbnb_cleaned,
    geometry=gpd.points_from_xy(df_airbnb_cleaned['long'], df_airbnb_cleaned['lat']),
    crs="EPSG:4326"
)

fig, ax = plt.subplots(figsize=(14, 12))

gdf.plot(
    column='number_of_reviews',
    ax=ax,
    legend=True,
    legend_kwds={'label': "Número de Reseñas"},
    cmap='viridis',
    markersize=gdf['number_of_reviews'] * 0.1,
    alpha=0.6
)

cx.add_basemap(
    ax,
    crs=gdf.crs.to_string(),
    source=cx.providers.CartoDB.Positron
)

ax.set_title('Distribución de Reseñas por Ubicación', fontsize=18)
ax.set_xlabel('Longitud')
ax.set_ylabel('Latitud')

plt.show()

# Definir las variables
numerical_features = [
    'lat', 'long', 'service_fee', 'minimum_nights', 'number_of_reviews',
    'reviews_per_month', 'review_rate_number', 'calculated_host_listings_count',
    'availability_365'
]

categorical_features = [
    'host_identity_verified', 'neighbourhood_group', 'neighbourhood',
    'cancellation_policy', 'room_type'
]

target = 'price'

#  X y y
X = df_airbnb_cleaned[numerical_features + categorical_features]
y = df_airbnb_cleaned[target]

# Train & Test
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Generar el pipeline de preprocesamiento
numerical_transformer = Pipeline(steps=[
    ('imputer', SimpleImputer(strategy='median')),
    ('scaler', StandardScaler())
])

categorical_transformer = Pipeline(steps=[
    ('imputer', SimpleImputer(strategy='most_frequent')),
    ('onehot', OneHotEncoder(handle_unknown='ignore', drop='first'))
])

preprocessor = ColumnTransformer(
    transformers=[
        ('num', numerical_transformer, numerical_features),
        ('cat', categorical_transformer, categorical_features)
    ])

# Lista de modelos a comparar
models = {
    'Regresión Lineal': LinearRegression(),
    'Árbol de Decisión': DecisionTreeRegressor(max_depth=10, random_state=42),
    'Bosque Aleatorio': RandomForestRegressor(n_estimators=50, max_depth=10, random_state=42),
    'Gradient Boosting': GradientBoostingRegressor(n_estimators=50, max_depth=5, random_state=42),
    'XGBoost': XGBRegressor(n_estimators=50, max_depth=5, random_state=42),
    'LightGBM': LGBMRegressor(n_estimators=50, num_leaves=31, random_state=42)
}

# diccionario para almacenar resultados
results = {}
predictions_for_plot = {}

# Entrenamiento y evaluación de cada modelo
print("--- Entrenando y evaluando modelos ---")
for name, model in models.items():
    print(f"Entrenando {name}...")
    start_time = time.time()
    
    model_pipeline = Pipeline(steps=[
        ('preprocessor', preprocessor),
        ('regressor', model)
    ])

    model_pipeline.fit(X_train, y_train)
    
    end_time = time.time()
    training_time = end_time - start_time
    
    y_pred = model_pipeline.predict(X_test)
    rmse = np.sqrt(mean_squared_error(y_test, y_pred))
    mae = mean_absolute_error(y_test, y_pred)
    r2 = r2_score(y_test, y_pred)
    
    
    results[name] = {'RMSE': rmse, 'MAE': mae, 'R2': r2, 'Tiempo': training_time}
    predictions_for_plot[name] = y_pred
    print(f"✅ {name} entrenado en {training_time:.2f} segundos.")

--- Entrenando y evaluando modelos ---
Entrenando Regresión Lineal...
✅ Regresión Lineal entrenado en 0.95 segundos.
Entrenando Árbol de Decisión...
✅ Árbol de Decisión entrenado en 3.05 segundos.
Entrenando Bosque Aleatorio...
✅ Bosque Aleatorio entrenado en 98.20 segundos.
Entrenando Gradient Boosting...
✅ Gradient Boosting entrenado en 19.27 segundos.
Entrenando XGBoost...
✅ XGBoost entrenado en 0.89 segundos.
Entrenando LightGBM...
[LightGBM] [Info] Auto-choosing col-wise multi-threading, the overhead of testing was 0.005397 seconds.
You can set `force_col_wise=true` to remove the overhead.
[LightGBM] [Info] Total Bins 1898
[LightGBM] [Info] Number of data points in the train set: 66576, number of used features: 178
[LightGBM] [Info] Start training from score 625.213335
✅ LightGBM entrenado en 0.89 segundos.

# Tabla de resultados
print("\n--- Comparación de Rendimiento de Modelos ---")
print("| Modelo                  | RMSE ($)   | MAE ($)    | R2 Score | Tiempo (s) |")
print("|-------------------------|------------|------------|----------|------------|")
for name, metrics in results.items():
    print(f"| {name:<23} | {metrics['RMSE']:.2f} | {metrics['MAE']:.2f} | {metrics['R2']:.2f} | {metrics['Tiempo']:.2f} |")

--- Comparación de Rendimiento de Modelos ---
| Modelo                  | RMSE ($)   | MAE ($)    | R2 Score | Tiempo (s) |
|-------------------------|------------|------------|----------|------------|
| Regresión Lineal        | 31.27 | 3.97 | 0.99 | 0.95 |
| Árbol de Decisión       | 17.41 | 1.76 | 1.00 | 3.05 |
| Bosque Aleatorio        | 13.67 | 1.64 | 1.00 | 98.20 |
| Gradient Boosting       | 14.16 | 2.36 | 1.00 | 19.27 |
| XGBoost                 | 16.00 | 2.07 | 1.00 | 0.89 |
| LightGBM                | 14.17 | 2.31 | 1.00 | 0.89 |

# Grafico de barras comparativo
metrics_df = pd.DataFrame(results).T
fig, axes = plt.subplots(nrows=1, ncols=3, figsize=(18, 6))
fig.suptitle('Comparación de Modelos por Métricas de Rendimiento', fontsize=18)

metrics_df[['RMSE', 'MAE']].plot(kind='bar', ax=axes[0])
axes[0].set_title('RMSE y MAE')
axes[0].set_ylabel('Error ($)')
axes[0].tick_params(axis='x', rotation=45)

metrics_df['R2'].plot(kind='bar', ax=axes[1], color='purple')
axes[1].set_title('R2 Score')
axes[1].set_ylabel('R2 Score')
axes[1].tick_params(axis='x', rotation=45)

metrics_df['Tiempo'].plot(kind='bar', ax=axes[2], color='teal')
axes[2].set_title('Tiempo de Entrenamiento')
axes[2].set_ylabel('Tiempo (s)')
axes[2].tick_params(axis='x', rotation=45)

plt.tight_layout(rect=[0, 0, 1, 0.95])
plt.show()

# Tabla comparativa de predicciones
sample_indices = np.random.choice(len(y_test), 10, replace=False)
sample_df = pd.DataFrame({'Precio Real': y_test.iloc[sample_indices].values.round(2)})

for name, y_pred in predictions_for_plot.items():
    sample_df[f'Pred. {name}'] = y_pred[sample_indices].round(2)
    sample_df[f'Error {name}'] = abs(sample_df['Precio Real'] - sample_df[f'Pred. {name}']).round(2)

print("\n--- Tabla de Comparación de Predicciones ---")
print("Muestra de 10 datos aleatorios del conjunto de prueba.")
print(sample_df.to_string())

--- Tabla de Comparación de Predicciones ---
Muestra de 10 datos aleatorios del conjunto de prueba.
   Precio Real  Pred. Regresión Lineal  Error Regresión Lineal  Pred. Árbol de Decisión  Error Árbol de Decisión  Pred. Bosque Aleatorio  Error Bosque Aleatorio  Pred. Gradient Boosting  Error Gradient Boosting  Pred. XGBoost  Error XGBoost  Pred. LightGBM  Error LightGBM
0        295.0                  298.60                    3.60                   295.11                     0.11                  295.01                    0.01                   296.43                     1.43     292.769989           2.23          296.56            1.56
1        626.0                  625.01                    0.99                   625.22                     0.78                  625.27                    0.73                   625.18                     0.82     622.940002           3.06          625.14            0.86
2        342.0                  343.58                    1.58                   339.96                     2.04                  340.08                    1.92                   341.48                     0.52     337.369995           4.63          341.51            0.49
3        358.0                  364.57                    6.57                   360.38                     2.38                  360.14                    2.14                   361.51                     3.51     360.239990           2.24          361.48            3.48
4        408.0                  414.50                    6.50                   409.78                     1.78                  409.63                    1.63                   411.06                     3.06     410.160004           2.16          411.06            3.06
5        417.0                  416.40                    0.60                   414.76                     2.24                  414.75                    2.25                   416.14                     0.86     412.929993           4.07          416.18            0.82
6        925.0                  925.05                    0.05                   925.31                     0.31                  925.27                    0.27                   923.57                     1.43     926.479980           1.48          923.27            1.73
7        192.0                  196.07                    4.07                   190.50                     1.50                  190.32                    1.68                   191.89                     0.11     192.399994           0.40          192.02            0.02
8        784.0                  784.16                    0.16                   785.47                     1.47                  785.04                    1.04                   784.04                     0.04     783.739990           0.26          784.09            0.09
9         78.0                   84.08                    6.08                    80.00                     2.00                   79.84                    1.84                    83.07                     5.07      80.720001           2.72           82.69            4.69

# Grafico de dispersión de predicciones vs valores reales
models_to_plot = list(models.keys())
num_models = len(models_to_plot)

if num_models <= 3:
    fig, axes = plt.subplots(nrows=1, ncols=num_models, figsize=(6 * num_models, 6))
else:
    nrows = (num_models + 1) // 2
    fig, axes = plt.subplots(nrows=nrows, ncols=2, figsize=(12, 6 * nrows))
    
fig.suptitle('Comparación de Predicciones por Modelo', fontsize=18)
axes = axes.flatten()

perfect_predictions = np.linspace(min(y_test), max(y_test), 100)

for i, name in enumerate(models_to_plot):
    ax = axes[i]
    ax.scatter(y_test, predictions_for_plot[name], alpha=0.5)
    ax.plot(perfect_predictions, perfect_predictions, color='black', linestyle='--', label='Predicción Perfecta (y=x)')
    ax.set_title(f'Predicciones vs. Reales: {name}')
    ax.set_xlabel('Precios Reales')
    ax.set_ylabel('Predicciones')
    ax.legend()
    ax.grid(True)
    
plt.tight_layout(rect=[0, 0, 1, 0.95])
plt.show()

Columna	Descripción
`id`	Identificador único de la propiedad en Airbnb.
`NAME`	Título o nombre del anuncio de la propiedad.
`host id`	Identificador único del anfitrión que publica la propiedad.
`host_identity_verified`	Indica si la identidad del anfitrión está verificada por Airbnb.
`host name`	Nombre del anfitrión.
`neighbourhood group`	Zona o distrito principal donde se ubica la propiedad.
`neighbourhood`	Barrio específico dentro del grupo anterior.
`lat`	Latitud de la ubicación de la propiedad.
`long`	Longitud de la ubicación de la propiedad.
`country`	País donde está ubicada la propiedad.
`country code`	Código del país (por ejemplo, `US` para Estados Unidos).
`instant_bookable`	Indica si la propiedad puede reservarse de forma inmediata.
`cancellation_policy`	Política de cancelación definida por el anfitrión.
`room type`	Tipo de alojamiento ofrecido: habitación privada, compartida o entera.
`Construction year`	Año de construcción del inmueble
`price`	Precio por noche del alojamiento (como texto con `$`).
`service fee`	Tarifa adicional por servicio cobrada por Airbnb.
`minimum nights`	Número mínimo de noches requerido para reservar.
`number of reviews`	Cantidad total de reseñas que recibió el alojamiento.
`last review`	Fecha de la última reseña recibida.
`reviews per month`	El promedio de reseñas mensuales que recibe el anuncio durante su vigencia.
`review rate number`	Puntuación promedio de las reseñas.
`calculated host listings count`	Cantidad de propiedades que tiene publicadas el mismo anfitrión.
`availability 365`	Días disponibles en el año para reservar la propiedad.
`house_rules`	Reglas de la casa especificadas por el anfitrión.
`license`	Número de licencia legal del alojamiento

EJERCICIO PRACTICO: SEGMENTACION DE ALOJAMIENTO Y PREDICCION DE PRECIO¶

Objetivo¶

Instrucciones Generales¶

Actividades a Realizar¶

Limpieza de datos:¶

Carga de datos y paqueterias a utilizar.¶

Tratamiento de Datos¶

TRANSFORMACION DE DATOS¶

Proceso de Eliminación de Columnas¶

Analizar tipos de datos en columnas objeto¶

Estandarizando los datos categoricos¶

Trabajando Duplicados y Datos Faltantes¶

Analisis de Outliers¶

Graficando Outliers¶

Calculo de Asimetria¶

Solucionar Asimetrias¶

Minimum nights¶

Capping¶

Resultados de Asimetrias¶

Analisis Exploratorio de Datos (EDA)¶

Correlacion¶

Variables Categoricas¶

Exploracion por Barrio¶

Graficos Geograficos¶

Conclusiones del EDA¶

Modelado Predictivo¶

Preparacion de los datos¶

Test¶

Pipeline de Preprocesamiento¶

Comparativa de Modelos¶

Comparativa de Pruebas¶

Conclusiones Finales del Proyecto¶

Recomendaciones de Negocio y Estrategias¶

	NAME	host id	host_identity_verified	host name	neighbourhood group	neighbourhood	lat	long	country	country code	...	service fee	minimum nights	number of reviews	last review	reviews per month	review rate number	calculated host listings count	availability 365	house_rules	license
id
1001254	Clean & quiet apt home by the park	80014485718	unconfirmed	Madaline	Brooklyn	Kensington	40.64749	-73.97237	United States	US	...	$193	10.0	9.0	10/19/2021	0.21	4.0	6.0	286.0	Clean up and treat the home the way you'd like...	NaN
1002102	Skylit Midtown Castle	52335172823	verified	Jenna	Manhattan	Midtown	40.75362	-73.98377	United States	US	...	$28	30.0	45.0	5/21/2022	0.38	4.0	2.0	228.0	Pet friendly but please confirm with me if the...	NaN
1002403	THE VILLAGE OF HARLEM....NEW YORK !	78829239556	NaN	Elise	Manhattan	Harlem	40.80902	-73.94190	United States	US	...	$124	3.0	0.0	NaN	NaN	5.0	1.0	352.0	I encourage you to use my kitchen, cooking and...	NaN
1002755	NaN	85098326012	unconfirmed	Garry	Brooklyn	Clinton Hill	40.68514	-73.95976	United States	US	...	$74	30.0	270.0	7/5/2019	4.64	4.0	1.0	322.0	NaN	NaN
1003689	Entire Apt: Spacious Studio/Loft by central park	92037596077	verified	Lyndon	Manhattan	East Harlem	40.79851	-73.94399	United States	US	...	$41	10.0	9.0	11/19/2018	0.10	3.0	1.0	289.0	Please no smoking in the house, porch or on th...	NaN

	NAME	host id	host_identity_verified	host name	neighbourhood group	neighbourhood	lat	long	country	country code	...	service fee	minimum nights	number of reviews	last review	reviews per month	review rate number	calculated host listings count	availability 365	house_rules	license
id
6026161	Upper East Side 2 bedroom- close to Hospitals-	65193709566	verified	Juliana	Manhattan	Upper East Side	40.76222	-73.96030	United States	US	...	$21	30.0	2.0	6/8/2019	0.21	3.0	34.0	157.0	NaN	NaN
6026161	Upper East Side 2 bedroom- close to Hospitals-	65193709566	verified	Juliana	Manhattan	Upper East Side	40.76222	-73.96030	United States	US	...	$21	30.0	2.0	6/8/2019	0.21	3.0	34.0	157.0	NaN	NaN
6026714	Close to East Side Hospitals- Modern 2 Bedroom...	31072202372	verified	Juliana	Manhattan	Upper East Side	40.76249	-73.96217	United States	US	...	$57	30.0	6.0	1/31/2019	0.14	3.0	34.0	67.0	The quieter the better, but otherwise make you...	NaN
6026714	Close to East Side Hospitals- Modern 2 Bedroom...	31072202372	verified	Juliana	Manhattan	Upper East Side	40.76249	-73.96217	United States	US	...	$57	30.0	6.0	1/31/2019	0.14	3.0	34.0	67.0	The quieter the better, but otherwise make you...	NaN
6027266	ACADIA Spacious 2 Bedroom Apt - Close to Hospi...	95854111798	verified	Juliana	Manhattan	Upper East Side	40.76021	-73.96157	United States	US	...	$117	30.0	10.0	11/18/2018	0.22	5.0	34.0	211.0	No Smoking No Pets	NaN
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
35606797	Bright and Beautiful Top Floor Two Bedrooms	65331079885	unconfirmed	Frankie	Brooklyn	Carroll Gardens	40.68383	-73.99281	United States	US	...	$205	30.0	18.0	11/4/2021	0.63	2.0	1.0	3.0	NaN	NaN
35607349	Modern & Bright Queen Bedroom Midtown East	57770451783	unconfirmed	David	Manhattan	Upper East Side	40.76132	-73.96064	United States	US	...	$28	30.0	1.0	11/4/2021	0.25	3.0	4.0	307.0	NaN	NaN
35607349	Modern & Bright Queen Bedroom Midtown East	57770451783	unconfirmed	David	Manhattan	Upper East Side	40.76132	-73.96064	United States	US	...	$28	30.0	1.0	11/4/2021	0.25	3.0	4.0	307.0	NaN	NaN
35607902	Modern NEW Room\|PRIVATE BATHROOM	7431680152	verified	Dariia & Jacob	Brooklyn	Bedford-Stuyvesant	40.68990	-73.94074	United States	US	...	$57	30.0	1.0	11/4/2021	0.25	2.0	10.0	365.0	NaN	NaN
35607902	Modern NEW Room\|PRIVATE BATHROOM	7431680152	verified	Dariia & Jacob	Brooklyn	Bedford-Stuyvesant	40.68990	-73.94074	United States	US	...	$57	30.0	1.0	11/4/2021	0.25	2.0	10.0	365.0	NaN	NaN